Distributed system evolution-some observations
ACM SIGOPS Operating Systems Review
Guardians and Actions: Linguistic Support for Robust, Distributed Programs
ACM Transactions on Programming Languages and Systems (TOPLAS)
Object management in distributed information systems
ACM Transactions on Information Systems (TOIS)
Resource management in a decentralized system
SOSP '83 Proceedings of the ninth ACM symposium on Operating systems principles
| Hi-index | 0.01 |
From the ConclusionsThe philosophy of language has indicated that there are seriousomissions in the Cartesian theory of categorisation; notably, thatmembers of a category are related by family resemblances and notnecessarily by shared properties between members of categories. Asobserved by Lakoff, these findings are also in conflict with otherpopular positions held in the Artificial Intelligence communitywhich view human reason as completely abstract, as well as assumingthat every cognitive process is algorithmic in nature; that thoughtis purely a matter of symbol manipulation [9]. The former of thesedoes not agree with findings of categorisation phenomena whichimply that basic level concepts are characterised by basic-levelperception and imaging capacities. The problem for the latterposition is that the only way it can make sense of theunderstanding of two different conceptual systems is by means ofthe translation of one into another, or the translation of bothinto a third system. Understanding in terms of experience does noteliminate such organisational differences, so this is a problemwhich needs to be addressed.Self-organising activity will involve, minimally, the managementof information, its representation both within a given organisationas well as across organisations and the effect of system evolutionon the object. The object model, as currently based on theCartesian view of categorisation, may not be sufficient for theserequirements. Instead, it may need to accommodate what have beentermed the holistic aspects of knowledge representation andcategorisation. In any event, in order to define a set ofrequirements for self-organising architectures, we will need tobear in mind not only the object, but also what is implied by theobject.